Re: How to count the number of notes in a .ly file? (David Kastrup)

[Urs Liska]

Am 11.12.2013 19:47, schrieb Jan Rosseel:
> > From: Urs Liska <address@hidden>
> >
> > I repeat my (corrected) suggestion which could be a workaround at
> least.
> > If you can use Frescobaldi from Git you can pull and checkout the
> > "accounting" branch (not "statistics") of my Frescobaldi fork on
> > https://github.com/uliska/frescobaldi. (Ah no, you don't necessarily
> need
> > Git, you can also download a ZIP file of any branch from Github) This
> allows
> > you to copy a rudimentary statistic to clipboard on a file basis (the
> open
> > document). This contains a line
> >
> > XX | Note
> >
> > which you could grep or something like this to get your numbers.
> >
> > Very hacky but _could_ work if the number of files is not too big.
> >
> > HTH
> > Urs
> I'll give that a try after our launch. It would be a good feature to
> have in Frescobaldi. It's a common feature in a word processor, and also
> belongs in a tool like Frescobaldi.

Yes, I intended to do something about it, but it got surpassed by so 
many other, more urgent tasks...
What I'm thinking of is a detailed statistics panel. This should count 
all sort of items: Notes, slurs, manual beams, function definitions, 
function calls etc.

Of course this has to consider a whole score including all includes, but 
(as you suspected correctly) currently it only works on a single file.
I'm thinking of an interface that shows you the include paths and allows 
you to stop the recursion at any point so libraries aren't accounted for.

A more fancy idea is to give different items a configurable relative 
weight (e.g.: note=10,articulation=15,\shape=40), and you can then 
assign a value to the weight units (e.g. 10 = 3Ct.) and get the money 
you can charge for your work.
You'd have to experiment for some time, calculating existing scores 
until they match the prize you consider appropriate (and realistic), so 
you could determine a way to (more) reliably charge engraving tasks 
based on the content instead of only number of pages.

Anybody interested in joining this is welcome (although I'm currently 
not planning to do any work on this).

Urs